Influence of Obstacles on the Prope1,ties ofa Medium.

481
limited to two particles only. Given the proper
with cprresponding conditions of mass , shape

temperature

, and distribu"'

tion of charge on the particles , and , as it seems to me , almost

any amount of molecular complexity is possible. That I have
not taken this possibility into account does not, however
vitiate the results here brolight forward , as they do not pre-

tend to greater accuracy than that of their order of magnitude.

It is the cumulative

value of these

results which wil,

hope, be regarded as suffcient reason for the publication of

what is at best an incomplete piece of theory.
Univ. CoIl. Bristol.

LVI.

On the Infuence of Obstacles arranged in

Order upon the Properties of a Medium. By

Rectangular

Lord RAYLEIGH

Sec. R.

HE remarkable formula , arrived at almost simultaneously

. by L. Lorenz t and H. A. Lorentz:j, and expressing the

relation between refractive index and density, is well known;

but the demonstrations are rather diffcult to follow,
limits of application are far from obvious. Indeed

and the

, in some

discussions the necessity for any limitation at all is ignored.

I have thought that it might be worth while to consider the
problem in the more definite form which it assumes when the
obstacles are supposed to be arranged in rectangular or square
order , and to show how the approximation may be pursued
when the dimensions of the obstacles are no longer very

small

in cOluparison with the distances between them.

Taking, first , the case of two dimensions , let us investigate
the conductivity for heat, or electricity, of an otherwise uniform
medium interrupted by cylindrical . obstacles which are arranged in rectangular order.

The sides of the rectangle wil

be denoted by IX , (3, and the radius of

the cylinders by

simplpst cases would be obtained. by supposing the

a.

The

material

composing the cylinders to be either non-conducting or per-

fectly conducting; but it will be suffcient to suppose that it

has a definite conductivity different from that ofthe remainder

of the medium.

By the principle of superposition

the conductivity of the

interrupted medium for a current in any direction can be deduced from its conductivities in the three principal directions.
Author.
xi. p. 70 (1880).
ix. p. 641 (1880).

. Communicated by the
t Wied.

Ann.

t Wied.

Ann.

JJprq &yl
Sip-ce CQ1J,gnction

igp.on

pl!rall

sl3nts nothing specilll

".
- ""I'

tl!e Influence pi Obstacles

l tQ the ax()s of the
for our consideration

cylinders pre, we may limit

(IX)

011r att Jltion to conduction parallel to one of the sides

thQ rectangular strqcture. In this case lin

13 parallel to

I;ig-. 1.

O't

o A

I0

symmetrically situated bet.ween the cylinders , such as AD

Be, are lines, of flow , and the perpendicuh1r lines AB , CD arc
equipotential
. If we take the centre of one of the cylinders P as origin of
polar coordinates ,

the potential external to the cylinder may

be expanded in the l3eri

V ==Ao

(AI1'+Bl

cos

0+

(Asr +Bs1'

) cos

30+..

" (n

and at point within the cylinder in the series

=CO+Cl1'CosO+Csr cos38+...

being measured from the direction,

of

(2)

IX, The sines of

and its multiples are excluded by the symmetry with respect

to

0=

, and the cosines of the even multiples by the. sym-

o==!w.
At the hounding
metry with respect to
we have the conditions
where

surface

r=q"

V =V'

vdV' /dr=dVjd1'

II denoting the condudiyity of

the material composing

the

cylinders in terms of tha'l of tpe remainder reckoned as unity.

The application of these

conditions to the

term in COB

gives

(3)

cylinders are perfectly conducting,

If they are non-c?nducting, 11=0.;

In the case where the

v=oo.

(!;)

. '. '

y)

II'

uodn puedep mM
unooou OW! uaJI'B~ aq O~ S! se;).IOS JO rna SAS
eHug.u! eq qo!lI U! JaUUl1m eq~ UBq~ ueAm:
pUll)' O~ IeTI'BIt
Ef
AJ'BpUnoq . 1'Bln:6uupaJ 'B aAuq: O~ s. mpUHAO aq~ Aq pe!dnooo

nO!~BJ (enuy !) eq~ eU!~'Bm! aM. J! SseUJ'Bep O~ eonpuoo TE

uI
,2'
Aq pB uep eq A'Bm F'Bd ~s.
PUOOBS Bq~ :6U!Jep!SUOO
q~ JO sex'B Bq~ UO pB 'Bn~!S seo.IOS eldmnm O~
AHug.U! ~'B seOJlOS pmJ:B

eql. ' S,lBp1m

PU'B (pesn'BO S! M.OF Bq~ qO!q M. Aq)

S'B pBp.1U~BJ eq

AULII ~u!od AU'B

u A rl1!~uB~od

eql.

neeM~eq uO!~'P!eJ ;WljJJ:J'B HJ!nbB.1 e,\\mn!pam

PU'B
pmp'B eq~ JO BS'BO eq~ JOd: 'AHun S! AHAnOUpuoo eq~ qO!qM.
JO mn!pem pa~dUJ.IB~U!Un Bq~ uodn JIO'Bq TI'BJ eM ' 0 = f a JI

(9)

nnse.r l'Bug.

O== f ij.!(;

+ IAFJ- or;

snq~ S!

eq; 'l

(v)

JO uO!~1:o!Iddu eq~ mO.

a.ug S! Iu.l~a~u! eq

JO ~Jtld S!q~

'Bq~ os

=1.p/AP

S09 (;-

SOO(I
eq~ O~ B

Y)=A

g +'Q

aH.\,\\ AUUI e,\\ snq;

SOD U! esoq~ eJ:'B nUseJ
nqp~noo mMqO!qA\ (1) IT! I3mJa~ AluO aq~ snq~ PU'B
SO\)-

=,1P/ilP- =uP/ilP '

SOD V=

Jno~nOO eq~ JO lJud J'BlnOJp eq~

av puu

ao ueeM~aq Su ' BI~uuwaJ; euo 01~U!puodsaJ.IOO I'B!Wfj~od U! II'BJ

S! sauH emus eq~ JBAO pu~e~u! eq~ JO

eq~ S! 1 A eJBqM. ' if A

JapU!'Bma.I aq~ JO BUI'B eql.

aq~ JO ~.r'Bd s!1~ JO anI'BA

u'B~oeJ eq1

ao uo '

OXI

snq~ S! ay

eql. ' 0

ua.rJ:o 1'B

SS0.10'B

up/ AP ' up/

qS!l'BA q~oq

' ao .IaAO 1'B.I~e~u!

Aq e ouBp A'Bm aM qo!qM

8P up/ APS

0~ eq1 s~uasaJdaJ

ay l oa SBPP! aq~ .WAO

ilP

SOD .t=X'=

amnssu aM J! ' il II!A pSI SU ' uo! 'Bn1a

ao'Bld'B'1 says!~ua A np!:aa.1 S!q~ U!q~!.M . d JBpn!IAo aq~
n'Bpa.I eq~ naaM.~aq UO!~a. 1 aq JO Jno~nOO eq~ 01

PU'B aoay

tv)

UP
A""

0=8p

up

ilP

QIe' IOaq~ s uaBJ:!) Aldd'B M on sn ~a'1

mUBJ~S aq~ uMop
-19 gn sSBd aM S1: 'JapU!IAO O~ J;apU!IAO mOJJ sap'B ' .1aAaMoq
Qy WJat ~S,Ig. Bql. ' Iu!~ua od JO eOJ:p.os eldmnm J'BI!l!S 'B S'B

p,1p,~a.1 eq A'B\l qouB puu ' s;wPU!IAO aq~ TI JOJ am'BS Bq~ AI!J'BS
"'saD

y 'r

tOJ'B ."

UInJp N

\1Qt

s~na!oHJaoG aq~ JO sanluA aqL

v uoan .t

tO ,1vZ;-l,fj1.V1af)

484

on the Influence of Obstacles

Lord Rayleigh

case , which suffces

for our purpose , is when we suppose the rectangular boundary
to be extended infinitely more parallel to IX than parallel to (3.
It is then evident that the periodic difference V may be

the shape of the rectangle. The simplest

HIX.
For the
H,x and equated to
difference due to the sources upon the axes wil be equivalent

reckoned as due entirely to

column at + X) , and the removal

of one at -00 , and in the case supposed such a transference is
immaterial*

to the addition of one extra

. Thus

(7)

=HIX .

simply, and it remains to connect H with BI'

This we may do by equating two forms of the expression,

near P. The part of the

potential due to H,xand to the Iqultiple sources Q (P not

for the potential at a point

included) is

Ao+AITCOS Of Asr

, if we subtract

H,x

s cos

30+,...;

we may say that the potential at

due to the multiple sources at Q is the real part of

(8)
iy)O
iy)
+As (x+iy)S +Ao
+ (AI H)(,x+
are the coordinates of' the same point wh8n reBut if
ferred to the centre of one of the Q' , the same potential may

-S

be expressed by
"" 1.

-1

,x

zy

s,x +

the summation being extended over all the Q'

the coordinates of a Q referred to P

=x-E, y
so that

=B

(,x

we obtain ,

iy-E- i'T)-n

(,x

rising powers of

+5B

+3B

=1.2.
and so on ,

'T

equating term to term

=Bl

1 . 2 . 3 , 4, . 5 Ao =

If E,

'T;

Since (8) is the expansion of (9) in

(x+iy),

(9)

'J,

+3.

1 ; 2 . 3 . 4 . 5 BI

(10)
6 + 3 . 4 . 5 , 6 . 7 Bo

8 + . . .

where
2,,

(g+i'T)-2n

(11)

the summation extending over all the Q'

. . It

would be otherwise jf the infiite

of another shape

e, fl'to be square.

rectangle were 8upposed to be

in Rectangular Order upon a :Medium. 485

. By, (3) each B can be expressed in terms of the correspond-

ing A. For brevity, we wil write

=va n'"
where

=(l +v)/(l-' v)..

(12)
(13)

Weare now prepared to find the approximate value of the

conductivity. From (6) the conductivity of the rectangle is

f!)

2'1Bl

IX 1. flVl

.I
l =

i!)

1-

IX

2'1Bl
IXf3H..

sO that the specific conductivity of the actual

currents parallel to IX is

medium for

2'1B

(14)

aflH' .

and the ratio of H to B l is given approximately by (10)

and (12).

In the first approximation we neglect I4'

As, A5 . . . Bs, B5 .. .. vanish. In this case
H=Al + B
+ I

6 . . . .,

so that
(15)

and the conductivity is

2 '1a

(16)

IXfl( v'

The second approximation gives

4,

- jT

(17)

and the series may be continued as far as desired,

The problem is thus reduced to the evaluation of the quan-

tities I ., I4' ' , .

We
wil
consider
first
the
important
particular case which arises when the cylinders are in
square
that is when
in (11) are then both
fl=lX.
E and
'f

order,

multiples of
I l!

IX , and we may write

-n 8

(18)

n =

where
= I

+im)-n

(19)

the summation being extended to all integral values of

positive or negative ,

m=O , m!=0.

except the pair

quantities 8 are thus purely numerical , and real.

The next thing to be remarked is that, since
,m

,m
The
are as

486

= (-

--- y=
" =

- --

=---. , .=(!(= -

--

.u_.-

tilL
Lord Rayleigh

on the Influence of Obstacles

much positive as negative , 8" vanishes for every

of
n.
This holds even when IX and /3 are unequal.

odd value
;11

Again
l +im) -2n

1)n
is odd

Whenever

im'+m)= - 8

", or

' +m)1)"8
vanishes. Thus for

sqnare order

"i-

(20)
= 810 = 8
O. .
Ths argument does not, without I'ese\'vation , apply to 8
sUm is not convergent; and the symmetry

In that case the

between
and
essential to the proof of evanescence , only
holds under the restriction that the infinite region over which

the summation takes place is symmetrical with respect to the

two directions IX and /3- , for example , square or circular.
On the contrary, we have supposed , and must of course continue

to suppose, that the region in question is infinitely elongated

iu the direction of IX.

The question of convergency

may be tested by replacing

the parts of the sum relating to a great distance by the corre-

sponding integral. This

(.x+

JJ 1,21

= rr

y)2n

os

2n()r dr dO

md herein
-2n+2/( -2n+2)

-2n+l dr

1 there is convergency, but if

= 1 the integral
'So that if
contains an infinite logarithm,
We have now to investigate the value of 8 appropriate to

our purpose; that is , when the summation extends over the

:!u, y= :!v

region bounded by.x=

infinite , but so that

where

are both

and

If we suppose that the tegion

v/u=O.

of summation is that bounded by.x= :tv y= :!v the sum

Fig,

vanishes by symmetry. We may therefore regard the summation as extending over the region bonnded externally by
:!v and internally by
iV=
:tv (fig. 2). 'Vhen
, the sum may be l"eplaced by the COl'esponding
:too

.'v=

is very great.

o I"

in Rectangula1' Order upon a JJIedium.

487

integral. Hence
d!Jdy
- 2

(21)

.x +

the limits for

Ultimately

and those for

being ::

!J

being

and 00

is to be made infinite.

We have
+v

v (.x+iy)2

.x + it.

.x-

and
2v d!J
l! +

= 2 tan

00 ""2 tan -I

l=i11,

AccordiI gly

(22)

7T.

In the case of square order ,

equations (10) (12) give

2-.
3 8'" 2
'"2

4 - --

'"8 -

1 '
'l

7Ta

., 8 8
VIX

= V
IX

7a

(23)

2 -

,"l 8S

and by (14)
Conductivity
If

=1- 27Ta

(24)

IX

the proportional

denote

cylinders

occupied by the

space

p=7Ta /1X2 ;

(25)

and
2p

ConductIvIty =

1-

(26)

2 2E 8 s8'l

4 -

Of the double summation

indicated in (19) one part can

be effected without diffculty.

Consider the roots of

im'1)=O,
sin
They are all included in the form

t:

s =

7T+zm7T

is any integer , positive ,

where

negative , or zero. Hence

may be written in the form

irn7T)

we see that sin (E

A 1-

1-

W17T

- '1

zm7T

tm'1

m7T

+ 27T

488

._--

Lord Rayleigh

--- -

-_.

- .

on the Influence of Obstacles

in which
A '=

-sin

im7r.

Thus
log 1cos

,-cot

im7r

sin E

+log

get

m7r

=log

1-

un7r

+ 7r

and add the two equations ,

If we change the sign of

sin

== log

log
Zln7r

SIn

+log J1-

we

1-m7r

+Iog f1- m7r-7r

m7r+7r

1.

whence, on expansion of the logarithms

sin
sin
4im7r + 3 sin im7r ....
2 sin
sin im7r

+ sin

1l

(im7r-7r):

im7r + 7r?

(im7r):

, 1

+ l.t J

(zm7r+7r)4 (

l(zm7r)4

H:6
+ 3S

m7r-7r

Z1n7r un7r+7r 6 +
6 +

By expanding the sines on t,he left and equating the correE, we find

sponding powers of

1 zm+
1 tm1 1
2 +

(1.m

un)
(zm

1)"

2 +...

zm+

27r

3.
1 27r
15 .
1)4

Z'n+

zm+

Sil

. 2'

zm7r

un7r

Sil

(27)

(28)

m7T

sm

SIn

zm7r

sm

. 7r

un7T

SIl

1)6

zm7r

. (29)

In the summation with respect tv m required in (19) we

are to take all positive and negative integral values. But in

we are to leave out the first term , corre-

m=O

the case ()f

=O.

sponding to

m=O

When

1 7r
sin im7r

= 3"'

(im)2

. in Rectangular Order upon a Medium.

which, as is well known , is the value of
12+

1)2 + 22+

489

2)2

Hence
m",'"

2'7
m=l

in

-2im'7

+!r

(30)

and in like manner

m=c.

2'7

- i si

m=1

2-6

im'7+sin

im'7t, .

(31)

m=oo

2'7

27 . 35

h\ sin- im'7

im'7+sin

We have seen

alr8ady that 8

im'7t. .

='7.

, and that

comparison of the latter with (30) gives

(32)

The

m=o:

(33)

m'7=
m=1

We wil now apply (31) to the numerical calculation of

, We

find:

sin -4imlT.

im1!.

- sin

'00749767

'0000562150

1395

Sum

00751165

'0000562152

so that

='7 x -03235020. .
In the same way we may verify (33), and that (32)
If we introduce

(34)

this value into (26), taking for=0.

example

the case where the cylinders are

get

= 1), we

non-conductive

1+p- '3058p

(35)

From the above example it appears that in the

summation
there isa high degree of convergency.

with respect to

490

Lord Rayleigh

on the Influence qf Obstacles

f A

The reason for tbis wil appear more clearly if we consider

the nature of the fiTSt summation (with respect to
(x+iy)-n where
(19) we have to deal with the sum of

receives the
for the moment regarded as constant
, instead of being concentrated at equidistant
x=m
, while

values

points , the values of

would become

were ulliformly distributed , the sum

+'" dx

(x+iy)
Now, n being greater than 1 , the value of this integral is
zero, We see , then , that the finite value of the sum depends
entirely upon the discontinuity of its formation , and thus a
high degree of convergency

when

increases may be ex-

pected.

The same mode of calculation may be applied without

diffculty to any particular case of a rectangular arraIlgement,

For

example , in (11)
' +im(3/a)a.+im(3)- =1X
leads , as
the summation with respect to
be given ,

If

7r

before , to

= sin2(im7Tf3/a.),

+irn(3/a-2

and thus

m=c.

. sin

=27T

(irn7T(3/IX)

m=l

+tW.

(36)

The numerical calculation would now proceed as before

and the final approximate result for the conductivity is given
by (16). Since (36) is not symmetrical

with respect to

and (3, the conductivity of the medium is different in the two

principal directions,
=7T,
And since this does
When (3 = IX , we know that
not differ much from !7T , it follows that the series on the right
of (36) contributes but little to the total. The same will be
provided the ratio of
true, even though

13

be not equal to

the tw.o quantities, be moderate. VIf e may then

with 7T , or with !7T

, if we are content

identify

with a very rough

approximation.
The question of the values of the sums denoted by
2n
*
are of the form

intimately conllected with the theory of

the O- functiolls

(7T/2K),

inasmnch as the roots of H(u), or

2mK+ 2m iK'
'" Qa.yley

' Ellptic Functions; ' p, 300, The notation is that of Jacobi.

in, Rectangula1' Order upon a Medium, 491

The analytical question is accordingly that of the expansion
of Jog

(.x)

in ascending powers of .x. Now

, Jacobi * has

himself investigated the expansion in powers of

,'v
7J) :: 21

lf'1

sin

IV- q9f4

sin 3.x

25/4

sin

(37)
(38)

5.x-

rrK' /K

were

So far as the cube of .7J

the

result is

=1-

3KE-(2-

)K2

! (39)

D being a constant which it is not necessary further to specify.

and E are

the elliptic functions of

usually so denoted.

By what has been stated above the roots of

form

71

(m+m iK' /K);

Bl(.x)

are of the

(40)

so that

=i1 3KE -(2-

+i11

the summation on the left being

values

ofm

and

m=O

except

P)K L . (41)
integral

extended to all

=O.

2' If K'

This is the general solution for

+im
sinc.e

=21 2KE

k2=!,

and

= 71

in general t,
EK'

+E'K- KK' =!71.

further it is convenient to use the form in

exponential factor is . removed from the series.

In proceeding

which an

This is

(.x)=Ale -lAB"

in which

.JA.x-Sl
A3
3T

A5:J A7
+S2

7T +

-Sa
51

(43)

B=

2fJ4

fJ, Sl =()fJ, S2, =fJ(1X

.J'
(42)

71 71

A=

So

the law of formation of

being
=2m(2m+ 1)fJ4

=1X/3(a

6fJ4

Sdsm/dfJ- 8fJ' dsm/dtX , (44)

Son + 1

'i '

Crelle ' Bd. 54 , p. 82,

' Ellptic Functions,' p. 49.

t Cayley

=!.

492

Lord Rayleigh

).

= -

! + ' . , -----

on the Influence of Obstacles

while
IX

k'2

(3=

(45)

v(kk'

Thus

I have thought it worth while to quote these expres-

sions , as they. do not seem to be

easily accessible; but I

propose to apply them only to t.he case of square order , K' = K

k'2

2(35

, s2

IX=O

AB=I/71

=(3,
and

(46)

(3=I/v2;

36(39

, S4

4 A

5" (47)

20 J
(m)

Hence
!1:

,,;2

. 0) (m)

10g
If

(48)

16. 35,

271

:ti\2' .,. are the roots of

:ti\.,

, we have

(m)/x=

4-

2-

51 ,

271

(m+im

Now by (40) the roots in question are

4 71
71

, 8

=71, 8

and thus

8A8

(49)

70.

in which
.3 .5
22 , 42 . 6

.3
K = 1 +

A =

' '2 +

?T

22 . 42

. 4 +

'8+...

= 1'18034,

Leaving the two- dimensional problem ,

to the case of a medium interrupted by

I wil now pass on

obstacles

spherical

arranged in rectangular order. As before , we may suppose

that the side of the rpctangle in the direction of flow is IX , the
two others being (3 and "1' The radius ofthe sphere is
The course of the investigation runs so nearly parallel
that it wil suffce to indicate some of the
that already given ,

expansions

steps with brevity, In place of (1) a d (2) we have the

V=Ao + (AIr + B
+(A +B" -n- )Y,,

(50)
(51)

Yl'r+... +

in Rectangular Order upon a Medium.

denoting the spherical surface harmonic of order

from the surface conditions

V=V'

vdV' /dn

493

And

dV/dn

we find

+v+
. In general

2n+I

8=n

We must now consider

Y,,

the limitations to be

e:,

cos

scp

'=0

+K

sin

(52)
imposed upon

scp),

(53)

where

= sinBO cos,,-sO

1)

(2n

l)

(54)

cOSn-S- 20

being supposed to be measured from the axis of

to
IX), and
from the plane of
cp
symmetry requires that
when

n=O)

tJ

(parallel

In the present application

tJz.

should be even , and that

(except
(71- 0) is written for

should be reversed when

Hence even values of

Further , no sines of

scp

are to be exclnded altogether.

are admissible. Thus we may take
(55)

=cos

=COS

O+H sin

0-.g- cos

- If

Y 5 = COS

cos

+ L sin

0 cos

(56)

2cp,

cos

+ ir cos

B (COS

O-t

cos

0)

2cp

cos

+ L 4 sin 0 cos cos 4cp.

(57)
In the case where /3=1 symmetry further requirl s that
, L =O,
(58)
In applying Green s theorem (4) the only difference is that
the area of the
surface
bounding

we must now understand by

the region of integration, If C denote the total current

flowing across the faces

/3ry, V

the periodic difference of

potential , the analogue of (6) is

, =0.

IXC-/3yV + 471B
(59)
We suppose , as before , that the system of obstacles , extended without limit in every direction , is yet infinitely more
than in the directions of /3 and ry.
extended in the direction of
Phil, Mag.

IX

S, 5, Vol. 34, No. 211.

1892. 2

Dec.

494

Lotd Rayleigh

on the Influence of Obstacles

if H.1; be the potential due to the sources at

HIX , and
other than the spheres

Then ,

infinity

. ;a

, VI=
f3ry

f1-

IX L

471

lXf3rH

so that the _ specific conductivity of the

parallel to

compound medium

IX is

471B
1lXf3ry

(60)

H.

We wil now show how the ratio BJ/H is to be calculated

approximately, limiting ourselves , howevel' , for the sake of
1X=(3= ry. The
simplicity to the case of cubic orde,' , where

potential round P , viz.

Aar Ya+ ,.
1"Y
may be regarded as due to Hx and , to the other spheres Q

acting as sources of potential. Thus , if we revert to rectangular coordinates and denote the coordinates of a point rela-.
, y, z and relatively to one of the Q' s by
tively to P by

, z we have

Ao+ (A

l'

H)x+Aa (J,

" x

in which

's

(61 )

=x-g, y

'T, z

=z-',

if
'T, s be the coordinates of Q referred to P. The left side
of (61) is thus the expansion of the right in ascending powers
did. v of
of x y, z. Accordingly, A - H is found by taking
member and then making x .y, z vanish, Iu

the right-hand

like manner- 6A wil be found

coeffcient, Now , at the origin

from the third differential

d ,,r- p

7- --'

d.'

in which
+ S2

=E2

It will be obsefved that we start with a harmonic of order

1 and that the differentiation raises the order to 2. The law
differentiation raises the order by unity is general;
and . so far as we shall proceed , the harmonics are all zonal

that each

JI::

in Rectangular Orde1'

upon a Medium.

'7V

495.

and may be expressed in the usual way as functions

where

fh=g/P.

Thus
d 0/
7J 1,

1"

(fh)

of

fh,

-3 P

In like manner

'7

tgp

5P
(fh)

and

d" x
dg"

= - 24

P-" P (fh).

The comparison of terms in (61) thus gives

H=(62)
In each of the quantities ,

the summation
is to be extended to all the points whose coordinates
are of
such as

the form

lOl , mIX , nIX

where

are any set of integers , positive or negative

except 0
,
0
,
O.
If
we
take
IX=
1
,
and
denote
the
corresponding
sums by 8 , 8 , . . , . , these quantities wil be
purely numerical , and
-n(63)

From (52) (62) we now

obtain

2+v a3
+28

321-

2a

410

Ct

(64)

which with (60) gives the desired result for the conductivity
of the medium.
We now . proceed to the calculation of 8 , We have

3fh

2g2 -rl-

s2

clg

By the symmetry of a cubical arrangement, it follows that

(rnp
(S2
for a space bounded by a. cube

so that if 8 were calculated

it would necessarily vanish. But for our purpose 8

by g=:t 00

be calculated over the space bounded

2M 2

is to

:tv

?J

496
!;= :t

Lord Rayleigh
where

have just seen ,

=::

::.

,\

on the Influence of Obstacles

is finally to be made infinite; and ,

as we

we may exclude the space bounded by

E=

::v,

'I= ::v

so that !82 wil be obtained from the space bounded by

g=v E=oo 'I= ::v
is suffciently great , tho summation

Now when

may be

replaced by an integration; thus

)agd'Id!;.
=- SSS
In this

00

+'l

a(j) dE=+u
-u

+'2

v d'I
2V
1= (V

+'l

) (2V

+'2i'

and finally
2d()

+u

-u (7,

+'2

(2V

+'2

0"; (2+tan

2ds

rl/ "'2

Jo ";(2-8
Thl:s
27T

(65)

= 3"
, If we neglect a

/lX

for the ratio of volumes

, and write

lZ.
47Ta

(66)

3(.
we have by

(1)0) for

the conductivity

(2+v)/(1-v)(2+v)j(1-v)+p
or in the particular case of

(67) *

(v=O)

non-conducting obstacles

(68)

. +2P ,

In order to carryon the approximation . we must calculate

&c, Not seeing any general analytical method , such as

was available in the former problem ,

.. Compare Maxwell'

I have calculated an
314,

' Electricity,'

(~

-.IP esaq Japu(l ' a.mp\l~S

Bq~ JO

(/. ---'

Fl

--

XI)

. ,

po!.wd aq~ qH

aAu \\ Bq ~'Bq~ pamnss1I

uospudmoo U! ~1Ja.
S! q
SB.Iaqds.1O SJapU!IAO pax!f pUB P!~!l A'q papn.qsqo
mn!pam snoas1l~ 'B U! UOnOm-BA'BM. JO marqold aq S! asaq~

n'

JO ~sa!dm!s BqJ, '

UO!

'BJq!A JO smarqold

U!'BpBO O~ pa!TddB

~nq ' mn!pBm punodmo;) u JO

Bq MOU Aum ABq

AHA!Pl1P

AH;)!Idm!s JO a1111s aq1

J01 passa.Idxa uaaq aA'Bq UO!~B~!~SaAU! JUO JO sHl1saJ Bq.L

-nOG aq~ su a 1In~uuI IU;)!lPBla U

1.1

1900,

1.1

9800, +

'1 ' I

'

I' '

I'

1.61:0,

I' ' I' '

I' ' I' '

1.600,

I: ' I '
I: ' I '
0'

ttIO, +

'1601.

I' 'I' '

1'(

1'900.

1090,

I: ' I' '

'1 ' 0 '
I: ' I' '
I: ' I' '

1:1

woo' +

9600, -

noo, -

9l.0.

1:171'0,

(69)

LL60'

I' ' I '

0'
I' '

I 'I '
I 'I '

96(H.
17601:,

T '0 '

0009, 1: +

S! nnSBJ aqJ:
uO!~'Bmums oq~ JO A~!nU!~UOOS!P aq~ uodn

~u!puadap Bnp!SaJ JO PU!11 U AIUO S! paJ!nbBJ mns Bq ~uq

a1BldmoG aq~ JaAO pa~u.I a~u! uaq.\\ saqS!UBA d
1Iq~ ~ouJ aq Jq pB.IlOA'BJ S! uO!1'Bm!xoJddu Bq~ JO ssaoons aql.

os ' aJBqds

0 +

gO+gg=
puooas aq~ U! PU'B

~1 +gg+~g=
~SJ9 Bq~ ul ' uo!~!sodUiOO JO spuPI OM.~ S1uasaJdoJ 6
.10
U!q~!"'10 a!I
dldm'Bxa JOJ ' asuo aq~ U! uom~BdaJ S!ql. ' 81
rp!qM s u!od II'B JOJ ~l nsaJ aq~ su!u~uoo alq'B ~U!AI.OnOJ BqJ,
PloJOM.~ ' qS!Ul!A OM.

pUB ' PIOj-IlOJ S! UO!~!~BdeJ f)q~ ' qS!UUA sapm!p.IOOO BB.Iql Bq~

JO auo.H ' samn 8 s! ~'Bq~ ' ~UU GO qO'Ba

w~ LWArt '

U! pa~uadaJ s! a1!ug

BJl! J '/1' qO!qM. JOJ m.

J '/1' JO sanl'BA OJBZ
PU'B B. I.!Hsod JO uO!~'Bep!suoo Bq~ o~ saAIBsJno ~!m!I Aum B
dg

+ f,

30ft'BlnmJoJ

Bq~ mOJJ uO!l'Bmmns pB.I!P Aq fS JO anI'BA a~um xoJddu

tO ,wznOUVliJ&'8 u

UlnJp&N v uodn .

L6v

u_-

498

on the Infuence of Obstacles

Lord Rayleigh

cumstances the flow of gas round the obstacles follows the

same law as that of electricity, and the kinetic energy of the
motion is at once given by the expressions already obtained.
In
fact the kinetic energy corresponding to a given total flow
is increased by the obstacles in the same proportion as the
electrical resistances of the original problem , so that the
influence of tbe obstacles is taken into account if we suppose

that the density of the gas is increased in the above ratio

resistances. In the case of cylinders in square order
(35), the ratio is approximately (I+p)/(l- p), and in the

of

case of spheres in cubic order by (68) it is approximately

(1+!p)j(l-

But this i.; not the only effect of the obstacles which we
considering the velocity of propagation. The potential energy also undergoes a change.
The space available for compression or rarefaction is now
(I-p) only instead of 1; and in this proportion is increased
the potential energy corresponding to a given accumulation
must take into account in

of gas , For cylindrical obstruction the square of the velocity

of propagation is thus altered in the ratio

I+p -7(Iso that if

f-

l+p

denote the refractive index,- referred to

that of

the unobstructed medium as unity, we find

I+p,
1)/p=constant

(f-

(70)

which shows that a medium thus constituted would follow

Newton s law as to the relation between refraction and
density of obstructing matter, The same law (70) obtains
also in the case of spherical obstacles; but reckoned absolutely the effect of spheres is only that of eylinders of halved
density, It must be remembered , however ,

that while the

velocity in the last case is the same in all directions , in the

case of cylinders it is otherwise, For waves propagated
parallel to the cylinders the velocity is uninfluenced by their
presence. The medium containing the cylinders has therefore some of the properties which we are accustomed to
associate with double refraction , although here the refraction
is necessarily single,

To this paint we shall presently return

meantime it may be well to apply the formulre to

the mQre general case where the obstacles have the properties of fluid , with finite density and compressibility,

but in the

*' Theory. of Sound

303,

, i

in Rectangula1' Ol'de1' upon a Medium.

499

To deduce the formula for the kinetic energy we hav(lonly

to bear in mind that density cortesponds to electrical

res.istance,
Hence , by (26), if
denote t.he density of the cylindrical obstacle , that of the remainder of the medium being
unity, the kinetic energy is altered by the obstacles in the

approximate ratio

(u+ 1)/(u- 1) +p
(u+ 1)/(a-- 1)The effect of this is the same as if the density of

(71)
the whole

medium were increased in the like ratio.

. The change in the

potential energy depends upon the

" compressibility " of the obstacles. If th

material com-

posing them resists compression m times as -much as the remainder of the medium , the volume
counts only as
p/m
and the whole space

p+p/m

available maybe reckoned as

instead of 1. In this proportion is the potential energy of a

given accumulation reduced, Accordingly, if p, be the refractiye index as altered by the obstacles
2 = (71) x

(l-p+p/m).

(72)

The compr6ssibilities of all actual gases are nearly the same

so that if we suppose ourselves to be thns limited ,

we may

set m= 1 , and

u+ 1)/(u- 1) +p
P, -

(u+ 1)/(u-

(73)

1)-

, as it may also be written

2 -

1 1

constant, .

(74)
we obtain in

In the case of spherical obstacles of density

like manner (m = 1)

(2u+1)/(u- 1)+p

(75)

+l)/(u- l)-

2 -1 1

In the general case ,

pressing

where

in terms of p,2

constant.
Tn

(76)

is arbitrary, the equation ex-

is a quadratic

simple formulre analogous to (74) and

, and there are no

(76),

It must not be forgotten that the application of these

formulro is limited to moderately smull values of

If it be

500

Lord Rayleigh

------

."

' "(..

on the Influence of Obstacles

desired to push the application as far as possible , we must

employ closer approximations to (26) &c, It may be remarked that however far we may go in this direction , the
final formula wil always give p.2 explicitly as a function of
For example ,

in the case of rigid cylindrical

obstacles ,

we

have from (35)

= (l-

It wil be evident

that

'3058p
(77)
'3058
results such as these afford no

foundation for a theory by which the refmctive properties of

addition from the corre-

a mixture are to be deduced by

sponding properties of the components, Such theories reoccurs in the first power only, as
quire formulre in which
in (76),

If the obstacles are themselves elongated , or , even though

their form be spherical , if they are disposed in a rectangular
order which is not cubic , the velocity of wave-propagation
becomes a function of the direction of the wavp,-normal. As
we may regard the character of the refraction as
in Optics ,

wave- surface.

determined by the form of the

The reolotropy of the structure wil not introduce

any cor-

responding property into the potential energy, which depends

only upon the volumes and compressibilities concerned. The
therefore , reduces itself to the consideration

present question ,

of the kinetic energy as influenced by the direction of wave-

propagation. And this , as we have seen , is a matter of the

electrical resistance of certain compound conductors , on the
supposition , which we continue to make , that the wavelength is very large in comparison with the periods of the

structure, The theory of electrical conduction in general

has been

treated by Maxwell (' Electricity,

parallel treatment of the present

297), A

question shows that in

all

cases it is possible to assign a system of principal axes

having the property that if the wave-normal coincide with

anyone of them the direction of flow wil also lie in the

same direction , whereas in general there would be a divergence, To each principal axis corresponds an effcient" den-

sity, "
in the,

and the equations of motion ,

applicable to the medium

, take the form

gross
do
do
if!;
17
=m1
=m1
' U'

dt2

=m1

where
1;, 17, S are the displacements parallel to the axes
the compressibility, and

d!; d17 d'

0:

o -

If X

, m,

in Rectangular 01'dm' upon a

fL, v

Medium. 501

are the direction-cosines of the displacement

of the. wave-nOl' mal , we may take

'T

where

fLO

i(lx+n1y+y.

vt)

Thus
do/dtV

&c.

lB(lX+mfL+nv),

and the equations become

JlV
from which ,

l(lX+mfL+nv),
mlm(lX+mfL+
mln(lX+mfL+
we get

on elimination of X :

fL

p m
if

, b,

+r?n

(78)

denote the velocities in the principal directions

The wave-surface after unit time

soid whose axes are

is accordingly the ellp-

, b , c,

As an example :

if the medium , otherwise uniform , be ob-

structed by rigid cylinders occupying a moderate fraction

of the whole space ,

parallel to

the velocity in the direction

the cylinders , is unaltered; so that

= 1/ (1 +
= 1
In the application of our results to the electric theory of
light we contemplate a medium interrupted by spherical , 01'

cylindrical ,

obstacles ,

whose inductive capacity is different

from that of the undisturbed medium. On the other hand

the magnetic constant is supposed to retain its value unbeing so , the kinetic energy of the electric

broken. This

currents for the same total flux is the same as if there were
no obstacles, at least if we regard the wave- length as in-

finitely great * . And the potential energy of electric displacement is subject to the same mathematical laws as the
resistance of our compound eleeirical conductor, specific
inductive capacity in the one question corresponding to

electrical conductivity in the ot.her.

Accordingly, if
denote the inductive capacity of the material composing the spherical obstacles , that of the undisturbed
medium being unity, then the approximate value of

* See Prof, Wilard Gibbs

Theories of Light, " Am,

" Comparison of the Elastic and Electric

Journ, Sci, xxxv, (1888).

= +

- = -,. -

. .

502 Influence of Obstacles on tlte Properties of a . jJfedium.

given at once by (67).

The equation may also be written in

the form given by Lorentz

+2p v+2 =constant;

(79)

.1,

and , indeed , it appears to have been by the above argument

that (79) was originally discovered.

The above. formula applies in

strictness only when the

spheres are arranged in cubic order * , and , further , when

is moderate. The next approximation is

(80)

fk

I p
If the

65

obstacles be cylindrical ,

v+4/3P

10/3

and arranged in square

order , the compound medium is doubly refracting, as in the

is supusual

leetlic theory of light , in which

the medium

posed to have an inductive capacity variable with

thedirectiun

of displacement , independently of any discontinuity in its

structure. The double refraction is of course of the uniaxal
kind , and the wave-surface is the sphere and ellipsoid of

Huygens,

For displacements parallel to the cylinders the resultant

inductive capacity (analogons to conductivity in the conduction problem) is clearly 1vp;

so that the value of
for the principal extraordinary index is

=1+(vgiving Newton

law for the

density.

(81)

l)p,

relation between index and

For the ordinary index we have

(26),
in which

= (1 +

v)/(l-

v),

while S4, S8' .,

have the values

&c. we get

given by (49), If we omit

(82)

fk

+ 1

(83)

+ 1

The g('l1eral conclusion as regards the optical application

is that ,

even if we may neglect dispersion ,

pect such f'ormulre as '

(79) to be more

we must not ex-

approximately
bodies,

than

correct in the case of dense fluid and solid

" An irregular isotropic arrangement would , doubtless, give the mme

result,

f:!